Ziqi Zhao

Jilin, China

Welcome to my homepage! I graduated from EPFL in 2023 as a computer science master student, and I am actively looking for Ph.D. opportunities.

My current research interests is trustworthy machine learning, with a focus on privacy-preserving federated learning. In general, I am also interested in adversarial robustness, model compression, graph learning, large language models, and WiFi sensing.

Last update time: 2024.6.7

Page views this month:

News

2023.04.10 Our paper FIS-ONE: Floor Identification System with One Label for Crowdsourced RF Signals is accepted to ICDCS 2023!

Education

École Polytechnique Fédérale de Lausanne (EPFL)

Master of Science in Computer Science

Computer Science - Specialization in Data Analytics

Master Thesis: Quantized Neural Networks for 6D Pose Estimation

September 2020 - Janurary 2023

GPA: 5.54/6, Transcript

École Polytechnique Fédérale de Lausanne (EPFL)

Exchange Program

Computer Science

Feburary 2019 - July 2019

GPA: 5.58/6, Transcript

Hong Kong University of Science and Technology (HKUST)

Bachelor of Science in Computer Science and Mathematics (Double Major)

Computer Science and Mathematics (General Math Track)

Bachelor Thesis: Indoor crowdsourced Wi-Fi fingerprinting with network embedding

September 2016 - June 2020

First class honor.
GPA: 3.62/4.3, Transcript

Research

Quantized Neural Networks for 6D Pose Estimation

École Polytechnique Fédérale de Lausanne (EPFL)

Master's Thesis supervised by Dr. Mathieu Salzmann in Computer Vision Lab. Score: 5.75/6.

Applied network quantization to 6D pose estimation models to improve efficiency.
Analyzed the sensitivity of different parts of the network against quantization.
Proposed a multi-stage quantization algorithm to improve 1%-5% accuracy for different models.

September 2022 - Janurary 2023

Pushing the Limits of Optical Character Recognition on Complex Multilingual Documents

École Polytechnique Fédérale de Lausanne (EPFL)

Semester Project supervised by Prof. Frédéric Kaplan in Digital Humanities Lab. Score: 5.75/6.

Finetuned Google Tesseract's OCR model for ancient Greek texts to improve overall recognition performance.
Decreased the character error rate (CER) of Greek texts from 4.55% to 1.97% and that of the whole commentary from 8.07% to 3.28%.
Developed an image pre-processing pipeline to further decrease the CER by 1%.

Feburary 2022 - July 2022

Network Pruning in Adversarial Training

École Polytechnique Fédérale de Lausanne (EPFL)

Semester Project supervised by Prof. Sabine Süsstrunk in Image and Visual Representation Lab. Score: 5.75/6.

Proposed an algorithm to find robust subnetworks from randomly-initialized binary networks based on Strong Lottery Ticket Hypothesis.
Designed an adaptive pruning strategy to balance subnetwork search space and expression power.
The binary subnetworks showed competitive performance among state-of-the-art full-precision ones and saved more than 30% FLOPs.
The work was extended and published on Advances in Neural Information Processing Systems (NeurIPS) 2022. See the work

Feburary 2021 - July 2021

Floor Identification using Crowdsourced RF Signals

Hong Kong University of Science and Technology (HKUST)

Research Project supervised by Prof. Gary Shueng Han CHAN.

Proposed a novel algorithm to efficiently and accurately identify a user's current floor using received Wi-Fi signals and limited labels.
Modeled Wi-Fi signals as a bipartite graph and applied network embedding to solve the missing value problem.
Achieved nearly 100% of identification accuracy and still have over 90% of accuracy when 70% of Wi-Fi access points are missing.
The work was published on the 42nd IEEE International Conference on Distributed Computing Systems (ICDCS) 2022. See the work

March 2020 - July 2022

Wi-Fi Geo-Fencing

Hong Kong University of Science and Technology (HKUST)

Research Project supervised by Prof. Gary Shueng Han CHAN.

Designed a novel and robust algorithm for identifying whether people go out of a predefined geo-fence.
Helped collect the data from various settings.
Part of the project work was published on IEEE Internet of Things Magazine Special Issue on Smart IoT Solutions for Combating the COVID-19 Pandemic. See the work

March 2020 - Janurary 2021

Indoor Crowdsourced Wi-Fi Fingerprinting with Network Embedding

Hong Kong University of Science and Technology (HKUST)

Bachelor's Thesis supervised by Prof. Gary Shueng Han CHAN.

Designed a novel algorithm to build up an indoor Wi-Fi fingerprinting system with high deployment efficiency.
Conducted extensive experiments to evaluate the performance of the system.

July 2019 - July 2020

Personalized Running Speed Estimation using Foot Inertial Sensor

École Polytechnique Fédérale de Lausanne (EPFL)

Semester Project supervised by Prof. Kamiar Aminian in Laboratory of Movement Analysis and Measurement. Score: 6/6

Implemented a running speed personalization model with machine learning.
Analyzed the major features that can contribute most to the model and implement several models for comparison.

Feburary 2019 - July 2019

Publications

Note: * indicates equal contribution.

FIS-ONE: Floor Identification System with One Label for Crowdsourced RF Signals

Proceedings of the 43rd IEEE International Conference on Distributed Computing Systems (ICDCS) 2023

Weipeng Zhuo, Ka Ho Chiu, Jierun Chen, Ziqi Zhao, S.-H. Gary Chan, Sangtae Ha, Chul-Ho Lee

[IEEE Page] [PDF] [Code]

July 2023

Robust Binary Models by Pruning Randomly-Initialized Networks

Advances in Neural Information Processing Systems (NeurIPS) 2022.

Chen Liu*, Ziqi Zhao*, Sabine Süsstrunk, Mathieu Salzmann

[OpenReview] [PDF] [Code] [Poster] [Slides]

September 2022

GRAFICS: Graph Embedding-based Floor Identification Using Crowdsourced RF Signals

Proceedings of the 42nd IEEE International Conference on Distributed Computing Systems (ICDCS) 2022

Weipeng Zhuo, Ziqi Zhao, Ka Ho Chiu, Shiju Li, Sangtae Ha, Chul-Ho Lee, S.-H. Gary Chan

[IEEE Page] [PDF] [Code] [Slides]

July 2022

IoT geofencing for COVID-19 home quarantine enforcement

IEEE Internet of Things Magazine Special Issue on Smart IoT Solutions for Combating the COVID-19 Pandemic, Vol. 3, No. 3, pp. 24-29

Jiajie Tan, Edmund Sumpena, Weipeng Zhuo, Ziqi Zhao, Mengyun Liu, S-H Gary Chan

[IEEE Page]

September 2020

Services

TBD

Hope I will have services to list here in the future. 😁

TBD

Work Experience

Machine Learning Engineer Summer Internship

Ketl.io, Switzerland

Scraped and parsed nearly 500,000 multi-lingual online documents.
Developed a translation and summarization service for legal documents written in Italian, French, German and English.

July 2021 - September 2021

Misc.

Work Experience

Machine Learning Engineer Summer Internship

Ketl.io, Switzerland

Scraped and parsed nearly 500,000 multi-lingual online documents.
Developed a translation and summarization service for legal documents written in Italian, French, German and English.

July 2021 - September 2021

Programming Languages & Tools

Proficient:

Familiar:

Languages

Chinese (Mandarin): Native Speaker
English: Proficient
German: Beginner
French: Beginner
Chinese (Cantonese): Beginner

Awards

University's Scholarship Scheme for Continuing Undergraduate Students, HKUST

2016, 2017, 2018, 2019
Reaching Out Award from Hong Kong Government, HKUST

2019
Dean's List in Engineering School, HKUST

2017 Spring
Provincial First Prize, National High Schools Physics Competition, Jilin Province, China

2015

Ziqi Zhao

News

Education

École Polytechnique Fédérale de Lausanne (EPFL)

École Polytechnique Fédérale de Lausanne (EPFL)

Hong Kong University of Science and Technology (HKUST)

Research

Quantized Neural Networks for 6D Pose Estimation

Pushing the Limits of Optical Character Recognition on Complex Multilingual Documents

Network Pruning in Adversarial Training

Floor Identification using Crowdsourced RF Signals

Wi-Fi Geo-Fencing

Indoor Crowdsourced Wi-Fi Fingerprinting with Network Embedding

Personalized Running Speed Estimation using Foot Inertial Sensor

Publications

FIS-ONE: Floor Identification System with One Label for Crowdsourced RF Signals

Robust Binary Models by Pruning Randomly-Initialized Networks

GRAFICS: Graph Embedding-based Floor Identification Using Crowdsourced RF Signals

IoT geofencing for COVID-19 home quarantine enforcement

Services

TBD

Related Conference Deadlines

NeurIPS 2023

ICLR 2024

IEEE S&P (Oakland) 2024

ICML 2024

USENIX Security 2024

CCS 2024

Work Experience

Machine Learning Engineer Summer Internship

Misc.